Scholarlink

<p>This study evaluates groundwater spring potential in the Penjween basin, a tectonically active region in northeastern Iraq, using four data-driven and knowledge-based models: Frequency Ratio (FR), Weight of Evidence (WoE), Shannon Entropy (SE), and Analytic Hierarchy Process (AHP). Fifteen influential factors, including lithology, lineament, TWI, SPI, DD, Pp, WTD, K, slope degree, length, aspect, altitude, plan and profile curvature, and land use, were analyzed to determine their impact on spring occurrence. An inventory of 128 springs was developed and randomly split into training (70%) and validation (30%) datasets. Results indicate that springs are primarily associated with permeable lithologies (Penjween Ophiolite, Siriginal Groups), high lineament density near faults, hill slopes, shallow water tables, and moderate SPI and TWI values. Most springs are located in bush, grazing, and agricultural lands, particularly in flat and valley areas. Model performance was evaluated using the ROC and SDI. The results showed FR and WoE with the highest accuracy (AUC = 0.794), followed by SE (0.648) and AHP (0.571), while AHP performed the weakest due to its reliance on subjective prioritization. The study concluded that FR and WoE are more applicable for spring potential mapping, as high spring densities in key zones confirm their predictive strength, offering valuable insights for groundwater management and planning in the Penjween basin.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A combined multi-statistical strategy for mapping groundwater spring potential in northeastern Iraq

  • 期刊论文

  • 发表日期：

    2025年10月21日

查看全文
  • Rebar Aziz Qaradaghy,
  • Diary Ali Al-Manmi,
  • Dara Faeq Hamamin

摘要

This study evaluates groundwater spring potential in the Penjween basin, a tectonically active region in northeastern Iraq, using four data-driven and knowledge-based models: Frequency Ratio (FR), Weight of Evidence (WoE), Shannon Entropy (SE), and Analytic Hierarchy Process (AHP). Fifteen influential factors, including lithology, lineament, TWI, SPI, DD, Pp, WTD, K, slope degree, length, aspect, altitude, plan and profile curvature, and land use, were analyzed to determine their impact on spring occurrence. An inventory of 128 springs was developed and randomly split into training (70%) and validation (30%) datasets. Results indicate that springs are primarily associated with permeable lithologies (Penjween Ophiolite, Siriginal Groups), high lineament density near faults, hill slopes, shallow water tables, and moderate SPI and TWI values. Most springs are located in bush, grazing, and agricultural lands, particularly in flat and valley areas. Model performance was evaluated using the ROC and SDI. The results showed FR and WoE with the highest accuracy (AUC = 0.794), followed by SE (0.648) and AHP (0.571), while AHP performed the weakest due to its reliance on subjective prioritization. The study concluded that FR and WoE are more applicable for spring potential mapping, as high spring densities in key zones confirm their predictive strength, offering valuable insights for groundwater management and planning in the Penjween basin.